R/R & Win Rate - Is It Good Enough

[earth_imperator]

If you had a strategy that has placed 1000 trades with a fixed R/R of 2.25:1 with an average win rate of 43.8%.

Is the strategy & return good enough, or worth trading, for you?

risk=1 ; reward=2.25 ; winRatePct=43.8 ; riskCalc=risk * (100 - winRatePct) ; rewardCalc=reward * winRatePct ; expectancy=rewardCalc / riskCalc ; riskCalc ; rewardCalc ; expectancy
56.2
98.55
1.7535587188612098
The reward for each risked $1 is $1.75.
So, it's a good strategy.

 

[Sprout]

I don't see how the metrics of this are bad, so I'm not sure why the negative feedback. If the win is $225, and the loss is $100, with a win rate of 43.8%, this looks like a damn good strategy. Even the low equity line on this simulation run is positive (if just barely). So essentially, there is a high chance of making decent money, and a very low chance of barely breaking even.

View attachment 323043

It appeals to folks with a high "intolerance to uncertainty" and builds a self-limiting habit. Sure if you are Risking $100 to make $225, it doesn't seem so bad. But until one gets enough data to see what their max drawdown would be, as one scales, their R rises. At some point trading becomes more about managing risk, protecting capital and being highly selective with the best setups for 1R at $100, $1000, $10k, $100k, $1m, etc. all have different psychological effects.

 

[earth_imperator]

It appeals to folks with a high "intolerance to uncertainty" and builds a self-limiting habit. Sure if you are Risking $100 to make $225, it doesn't seem so bad. But until one gets enough data to see what their max drawdown would be, as one scales, their R rises. At some point trading becomes more about managing risk, protecting capital and being highly selective with the best setups for 1R at $100, $1000, $10k, $100k, $1m, etc. all have different psychological effects.

IMO the best mathematical solution as well a comforting psychological way is to divide your capital into 100 parts, each part being a trade (ie. investment). This means your max risk per trade is just 1% of your account...

Of course the max risk of the trade itself must not be above 100% of the investment (as is possible when shortselling etc.). Ie. the max risk must be known in advance by simply computing it with math tools like this one.

This of course is a kind of diversification and risk management. Ideally each ticker should not be used more than once at any time.

With small accounts one can adjust the amount per trade to say max 5% of the account, ie. then dividing the account value into 20 parts instead... I would not recommend to risk more than 5% of the account in any single trade.

Of course that many trades at the same time one can manage only by doing it programmatically.... (ie. program trading / auto trading / bot trading).
Some innovative trading companies or brokerages should offer such advanced trading services...
If a venture capitalist would like to invest into such a project idea and company I would be interested in taking part as an architect & developer & programmer in the project.

 

[ESHMUZ23]

Sorry for the delay, my posts were waiting mod approval. For those that did not see, the draw down is 33.6%

I agree with Noah. (Pretty rare )
I have such a system and I'm happy with the performance.
What you have to give us now is the rest of the story.
Such as what risk, position size, and number of trades.
Day, swing or position trading?

The # of trades were given in original post, 1k back tests.

Aside from psychological, the position size for the strategy is basically irrelevant due to the fixed RR. You can use $200/$450 as an example though. You can imagine your own scaled scenario to fit your budget to determine if the numbers would be good enough for you to run.

 

[deaddog]

Sorry for the delay, my posts were waiting mod approval. For those that did not see, the draw down is 33.6%



The # of trades were given in original post, 1k back tests.

Aside from psychological, the position size for the strategy is basically irrelevant due to the fixed RR. You can use $200/$450 as an example though. You can imagine your own scaled scenario to fit your budget to determine if the numbers would be good enough for you to run.

1000 trades over what time period?
You didn't mention the trading time frame or how much of the portfolio was at risk per trade.

 

[metalztrader]

To say this is bad is absurd. If one really believes they have a betting system that pays $225 for a win, loses $100 with a loss and wins 43.8% of the time then the ev of single bet is $42.35 . The kelly fraction is 20%.
You could start with $700 and print money until you retire. Or be conservative and start with half kelly and $1400. All fantasy land stuff.

I don't see how the metrics of this are bad, so I'm not sure why the negative feedback. If the win is $225, and the loss is $100, with a win rate of 43.8%, this looks like a damn good strategy. Even the low equity line on this simulation run is positive (if just barely). So essentially, there is a high chance of making decent money, and a very low chance of barely breaking even.

View attachment 323043